Download Assign Overpayment to Insurance Data with Adjustments

Paper GH-13
Assign Overpayment to Insurance Data with Adjustments
Qiling Shi, NCI Information Systems, Inc., Nashville, Tennessee
ABSTRACT
The purpose of this study is to assign overpayment to insurance data with adjustments using SAS®. Suppose we have
an algorithm to calculate the overpayment for each insurance ID, we need to join back with the original data to get
the adjustments and assign the overpayments to the insurance claims with adjustments. Since the overpayments are
calculated by rolling up the original dataset, we need to join back to get the adjustments by the same insurance ID.
There are some rules we need to consider to assign the values of overpayments after this join back. A SAS macro
with procedures such as PROC SQL and DATA STEPS is employed to do the data analysis.
INTRODUCTION
An insurance Claim is filed by a policyholder stating that an insured event has occurred and that the insurance
company should provide coverage. An insurance adjuster could be set up with access to portions of the claim to just
the right payment amount.
Insurance adjustment may be very complicated. For example, claims that be rejected as duplicates may be adjusted,
cancelled or resubmitted. Claims that rejected for eligibility or other billing errors may be adjusted when the eligibility
or billing issue is resolved. At different times, billing transactions that have previously processed may be paid or
repaid with a duplicate payment.
The Division of Medicaid and the fiscal agent allow adjusting and voiding of claims. If a paid claim is being adjusted,
the Provider Identification Number and the Recipient cannot be changed. The voided amount originally paid will
appear as a negative amount and that amount will be deducted from payments until the overpayment is recovered.
Usually we identify the overpayments for each insurance ID by different models or various definitions of rule
violations. In most cases, we need to roll up the original data by the insurance ID and then calculate the
overpayments. Hence we get a tuple with the information for overpayments according to each insurance IDs. But we
need some other information besides overpayments such as adjustments, dates of services and service codes etc.
In order to get this kind of information, for example the adjustments, we will join the overpayment tuple back with the
original data. To avoid the inflation of the overpayments and confusing, we will assign the overpayments to the
original claims which have adjustment indicators equaling zero or to the first claims with positive adjustments if there
are no original claims found or the original claims have been voided. For the other claims, we will assign the
overpayments as missing.
PROGRAM CODE
***********************************************************************************;
*
1. Join back with the original data to get the adjustments
;
*
2. When the adjustments in (0,2,4) and not the first id
;
*
by sorting, assign the overpay amount. Otherwise assign
;
*
the missing value as overpay.
;
***********************************************************************************;
%macro joinadj(old_data, new_data, keepvar, dataout);
*Join back with the original data to get the adjustments;
proc sql;
create table join_data as
select a.*, &keepvar.
from &old_data a, &new_data. b
where a.id = b.id;
quit;
1
*Create a new sorting ID: temp_indicator;
data new_sort;
set join_data;
if adjustment_indicator in (0,2,4) then temp_indicator =0;
else if adjustment_indicator =3 then temp_indicator =1;
run;
*Sort the data by id, temp_indicator adjustment_indicator;
proc sort data=new_sort;
by id temp_indicator adjustment_indicator;
run;
*Assign the overpay as missing to non-first IDs by the sorting order;
data &dataout.;
set new_sort;
by id temp_indicator adjustment_indicator;
if not first.id then overpay=.;
run;
%mend joinadj;
*Read in the insurance data;
data insurance;
infile "C:\Documents and Settings\shiq\Desktop\New Folder\NESUG\joinadj.txt";
input id $6. amount_charged adjustment_indicator;
run;
* Calculate the overpayments;
proc sql;
create table flag as
select id, sum(amount_charged) as overpay
from insurance
group by id
having overpay > 0;
quit;
*use the macro to get back the adjustments and assign overpay;
options symbolgen mprint mlogic;
%joinadj(insurance, flag, overpay, new);
RESULTS
The demonstrated data set “INSURANCE” has 18 observations and 3 variables. One variable is called “ID” which
represents the insurance claim identifications. “AMOUNT_CHARGED” means the provider charged on the insurance
agents or companies. “ADJUSTMENT_INDICATOR” provides codes for different adjustment activities. Here the code
“0” represents the original claims. The codes “2” and “4” mean claims with positive partial adjustments. The code “3”
indicates the claims with negative partial adjustments.
Table 1: The Dataset “INSURANCE” used as an example.
2
For the same claim ID, there are different records with different adjustments. We want to find out the overpayments
for each insurance ID.
The following is the table derived from “INSURANCE” containing the overpayment for the corresponding insurance
ID. Here we calculate the overpayments as the summation of “AMOUNT_CHARGED” grouped by “ID”.
Table 2: Dataset “FLAG” used to check voided claims for the same claim IDs.
We only keep the insurance ID with positive overpayments. That means removing those insurance IDs which do not
have overpayments or the overpayments are negative. There are 6 distinct insurance IDs having positive
overpayments identified.
Table 3: Dataset “JOIN_DATA”.
3
From table 3, we can get all the claims with positive overpayments after join the table “FLAG” back with the table
“INSURANCE” by “ID”. But in table 3, the amount of overpayments has been inflated. Originally we only have $257
overpayments, but now we have $617 overpayments. The problem is that we assign the overpayments to all the
claims. For example, for insurance ID “100001”, the overpayment $60 has been assigned three times to different
adjustments. To avoid this problem, we should only assign the overpayments to the original claims with adjustment
indicator equaling zero. If there is no original claim for this insurance claim ID, we will assign the overpayments to
the claims with positive adjustments.
Table 4: Dataset “NEW_SORT”.
We create a new sorting ID called “TEMP_INDICATOR”. If ADJUSTMENT_INDICATOR equals 0, 2 or 4, then
TEMP_INDICATOR =0. Otherwise if ADJUSTMENT_INDICATOR equals 3 then TEMP_INDICATOR =1. Then first
we sort the dataset by “ID”. Within each ID group, sort the dataset by “TEMP_INDICATOR” in an ascending order.
So it will put claims with ADJUSTMENT_INDICATOR equals 0, 2 or 4 before the claims with
ADJUSTMENT_INDICATOR equals 3. Within the group with ADJUSTMENT_INDICATOR equals 0, 2 or 4, sort the
dataset by ADJUSTMENT_INDICATOR in an ascending order. So the first claim should have
ADJUSTMENT_INDICATOR equals 0. If there is no original claim for a single insurance ID, the first claim should
have ADJUSTMENT_INDICATOR equals 2. For example, for insurance ID “100002” and “100007”, the first claim
has ADJUSTMENT_INDICATOR equals 2.
Table 5: Dataset “NEW”.
4
From table 5, we know that we assign the overpayments only to the original claims. If there is no original claim for
the insurance ID, for example, for insurance ID “100002” and “100007”, we assign the overpayments to the claims
with ADJUSTMENT_INDICATOR equaling 2. If there is no claim with ADJUSTMENT_INDICATOR equals 2, the first
claim should have ADJUSTMENT_INDICATOR equals 4. The total overpayment is $257 which is exactly as we
calculated for the dataset “FLAG”.
DISCLAIMER
All opinions and suggestions stated in this paper do not necessarily reflect the opinions and suggestions of NCI
Information Systems, Inc.
REFERENCES
1.
“Detecting Medicaid Data Anomalies Using Data Mining Techniques”, Southeast SAS Users Group
Conference, 2010.
2.
“Find Potential Fraud Leads Using Data Mining Techniques”, Southeast SAS Users Group Conference,
2011.
3.
“Remove Voided Claims for Insurance Data”, Western Users of SAS Software Conference, 2011.
CONTACT INFORMATION
Your comments and questions are valued and encouraged. Contact the author at:
Qiling Shi, Mathematics PhD, Certified Fraud Examiner
NCI Information Systems, Inc.
2636 Elm Hill Pike, Suite 115
Nashville, TN, 37214
Email: [email protected] Email: [email protected]
Website: http://www.linkedin.com/in/qilingshi
SAS and all other SAS Institute Inc. product or service names are registered trademarks or trademarks of SAS
Institute Inc. in the USA and other countries. ® indicates USA registration.
Other brand and product names are trademarks of their respective companies.
5